å°å·ãã ã¡ã¼ã«ã§éã ããã¹ã HTML é»åæ¸ç± PDF ãã¦ã³ãã¼ã ããã¹ã é»åæ¸ç± PDF ã¯ãªããããè¨äºãMyãã¼ã¸ããèªããã¨ãã§ãã¾ã Google DeepMindã®ç 究è ãã¯ç±³å½æé11æ17æ¥ããReinforcement Learning with Unsupervised Auxiliary Tasksãã¨é¡ããæ°ããªè«æã®å ¬éãå社ããã°ã§çºè¡¨ãããè«æã«ã¯ãåç©ã夢ãè¦ããã¨ã§å¦ç¿é度ãåä¸ããã¦ããç¹ããçæ³ãå¾ãããUNsupervised REinforcement and Auxiliary Learningãï¼UNREALï¼æ師ãªãã®å¼·åããã³è£å©å¦ç¿ï¼ã«ã¤ãã¦è¨ããã¦ããã ãã®ç¥è¦ãåãå ¥ããã¨ã¼ã¸ã§ã³ãããAtariã®ã²ã¼ã ã¨ããLabyrinthãã¨ãã3Dè¿·è·¯ã²ã¼ã ã使ã£ã¦ãã¹ãããã¨ããã以åã®ã¢ã«ã´ãªãºã ã«æ¯ã¹ã¦å¦ç¿é度ã10åã«é«
{{#tags}}- {{label}}
{{/tags}}